The Development of Autocatalytic Structural
Materials for Use in the Sulfur-Iodine Process for
the Production of Hydrogen
by
Kevin Miu
Submitted to the Department of Mechanical Engineering
and
Nuclear Engineering
in partial fulfillment of the requirements for the degrees of
Bachelor of Science in Mechanical Engineering
and
Nuclear Engineering
at the
MASSACHUSETTS NSTITUTE.OF TECHNOLOGY
may 2006
( Massachusetts Institute of Technology 2006. All rights reserved.
Author
......... ..
....
D..................................
Department of Mechanical Engineering
and
Nuclear Engineering
/,
Certifiedby.
May 12, 2006
A
...
k f
.........................................
Ronald G. Ballinger
Science and Engineering
Nuclear
rL _!...-
IieS:1
6upervisor
Accepted
n H. Lienhard V
.
'.k
LP.
J
.
.118k
.L
A.1q.'.J.A.L
inal rnefinparing
.kA.,CA.
..
k.
&.k.A...I.A.
Chairman, Undergraduate Thesis Committee
MASSACHUSETTS INSTITUTE.
OF TECHNOLOGY
AUG 0 2 2006
LIBRARIES
ARCHIVES
The Development of Autocatalytic Structural Materials for
Use in the Sulfur-Iodine Process for the Production of
Hydrogen
by
Kevin Miu
Submitted to the Department of Nuclear Engineering
and
Mechanical Engineering
on May 12, 2006, in partial fulfillment of the
requirements for the degrees of
Bachelor of Science in Nuclear Engineering
and
Mechanical Engineering
Abstract
The Sulfur-Iodine Cycle for the thermochemical production of hydrogen offers many
benefits to traditional methods of hydrogen production. As opposed to steam methane
reforming - the most prevalent method of hydrogen production today - there are no
carbon dioxide emissions. Compared to other methods of hydrogen production, the
efficiency of the cycle is excellent. Due to the high temperatures necessary for the
cycle, which are generally greater than 8500 C, several of the Generation IV nuclear
reactor concepts are attractive thermal energy sources. However, the high temperature and corrosive reaction conditions of the cycle, involving reactions including the
decomposition of H2 SO4 at 400-9000 C, present formidable corrosion challenges. The
conversion of sulfuric acid to sulfur dioxide was the focus of this study. The alloying
of structural materials to platinum has been proposed as a solution to this problem.
A catalytic loop to test the materials was constructed. Sulfuric acid was pumped over
the material at 903+20 C. The sulfur dioxide production of the catalyst was measured
as a means of quantifying the efficiency of the system as a function of temperature.
The maximum possible production of the material was calculated by using a mass
balance. A gas chromatograph was used to calculate the actual production of sulfur
dioxide. The results of the experiment show that an molecular conversion efficiency
of 10% is attained when operating at 900C while using 800H + 5%Pt as a catalyst.
The research confirms the catalytic activity of the material.
Thesis Supervisor: Ronald G. Ballinger
Title: Professor of Nuclear Science and Engineering
3
4
Acknowledgments
I would like to acknowledge Professor Ballinger firstly, because over the last three
years that I have known him, I have grown tremendously as both academically and
personally. I have, and I will be forever indebted to him for his patience and mentorship.
I would also like to thank Pete Stahle, Jeongyoun Lim, Jeff Hixon, and Monica Ware,
who spent countless days taming experimental setups, furnaces, and chemicals with
me in lab. Pete has always been a supreme resource in terms of machining knowledge, and I am thankful for the chance to learn even a small fraction of his craft.
Jeongyoun has provided invaluable help in the construction and testing of the experiment and knowledge about just about everything. Jeff was always willing to offer a
hand whenever a specimen had to be cut or parts had to be assembled. Monica was
instrumental in getting this experiment off the ground. Whether it was machining
a vital part or taking a trip to Anna's Tacqueria, all their friendship and company
were greatly helpful and valued.
Additionally, I owe much gratitude to Barbara Keesler, who helped tremendously by
handling all administrative duties. She was always there when I needed a friendly
face to talk to or to discussthe latest newsand sports.
Lastly, I must thank my family, without whose support, I would have never survived
MIT. They have watched their brother or son grow up, and I hope to continue to
honor them in the future.
5
6
Contents
1 Introduction
13
1.1 Motivation for Hydrogen Production
.....
1.1.1
Directions in Hydrogen Production
1.1.2
The Nuclear-Hydrogen Initiative
13
. .
....
1.2 Methods of Production ...........
1.2.1
Steam Methane Reforming .......
1.2.2
Electrolysis
1.2.3
High Temperature Steam Electrolysis .
1.2.4
Sulfur-Iodine Cycle ..........
1.2.5
Summary of Methods ..........
...............
1.3 Difficulties in Production ............
1.3.1
Thermal Considerations.
1.3.2
Structural Considerations.
1.3.3
Past Research ..............
1.4 Material Selection ................
1.5 Determination of Catalytic Ability
......
.............
.............
.............
.............
.............
.............
.............
.............
.............
.............
.............
.............
.............
.............
2 Methodology
2.1
System Requirements .......
2.2
Loop Construction.
2.3
Reaction Vessel Design ......
2.4
Hardware Interfaces ........
2.5
Hardware Settings.
15
15
16
16
17
18
19
21
22
22
23
23
24
25
27
....................
....................
....................
....................
....................
7
27
28
30
33
35
2.6
Material Preparation
...........................
37
2.7
Material Installation
...........................
37
2.8
Chromatograph Setup
38
..........................
2.8.1
Chromatograph Calibration
2.8.2
Valve Timing
40
...................
42
...........................
3 Results
3.1
3.2
43
Mass Flow Rate Analysis . . . . . . . . . . . . . . . .
.
43
......
3.1.1
Mass Flow Rates at Steady State Temperature .........
44
3.1.2
Mass Flow Rates as a Function of Temperature
.......
46
48
Gas Chromatograph Analysis ......................
3.2.1
Gas Chromatograph Analysis During Steady State Temperature 49
3.2.2
Gas Chromatograph Analysis as a Function of Temperature
3.3
Efficiency Calculations
..........................
3.4
Discussion and Conclusions
.......................
8
.
56
59
61
List of Figures
1-1 Electrolysis Cycle .......................
18
1-2 Sulfur-Iodine Cycle.
20
1-3 Temperature Dependence on Efficiency of the Sulfur-Iodine Cycle
.
23
2-1
Ismatech Micrometering Pump
2-2
Piping and Instrumentation Diagram .......
30
2-3
Reaction Chamber, Units in Inches ........
31
2-4
Mettler Toledo Scale ................
34
2-5
179A Flow Meter ..................
34
2-6
6890N Agilent Technologies Gas Chromatograph .
35
2-7
Dryer Connected to Compressed Air
35
2-8
Catalytic Loop ...................
36
2-9
Flow Readout Panel
36
..........
28
.......
................
2-10 HP Data Acquisition System ............
37
2-11 Catalyst Sample in Reaction Vessel ........
38
2-12 Valve Layout
39
....................
2-13 Nitrogen Calibration Curve
............
2-14 Sulfur Dioxide Calibration Curve
40
.........
41
3-1 Temperature and Mass Flow Data ................
3-2 Steady State Temperature and Mass Flow Data
44
........
45
3-3 Mass Flow Data with Dynamic Temperature ..........
47
3-4 Unaccounted Mass Flow Rate in Respect to Temperature . . .
48
3-5 Chromatograph Analysis at 5 Minutes of Steady State Period
49
9
3-6 S02 Peak at 5 Minutes of Steady State Period
3-7 Steady State Nitrogen Mass Flow Rate Data
.
.
............
50
.............
52
3-8 Chromatograph Analysis at 25 Minutes of Steady State Period ....
54
3-9 SO2 Peak at 25 Minutes of Steady State Period
54
.
...........
3-10 Chromatograph Analysis at 5950C ...................
3-11 SO 2 Peak at 595C
. . . . . . . . . . . . . . . .
.
.
.........
56
57
3-12 SO2 Production
..............................
59
3-13 SO 2 Production
Efficiency
60
........................
3-14 Catalyst Sample After Experiment
10
.
..................
61
List of Tables
2.1 Fluid Constants ..............................
2.2
32
Integration Results of Peaks of Nitrogen Calibration
.
.........
40
......
41
2.3 Integration Results of Peaks of Sulfur Dioxide Calibration
2.4 Calibration Values
............................
2.5 Valve Timing and Detection Events
42
42
..................
.
..
49
3.2 Integration Results of Peaks at 5 Minutes of Steady State Period ...
51
3.1
Composition
of Incoloy 800H . . . . . . . . . . . . . .
.
. .
3.3 Integration Results of Peaks at 25 Minutes of Steady State Period
3.4
Integration Results of Peaks at 595°C
11
.
.................
.
55
57
12
Chapter 1
Introduction
1.1
Motivation for Hydrogen Production
The viability of nuclear energy for hydrogen production has increased dramatically
in recent years. Nuclear energy answers two major concerns of hydrogen production:
climatic change due to greenhouse gases and the instability of fossil fuel supplies [1].
Generation IV nuclear reactors provide the requisite temperatures necessary for high
efficiency hydrogen production using the Sulfur-Iodine Cycle; depending on the reactor temperature, the efficiencies can be over 60% [2]. Before nuclear energy can be
deployed for large scale and long term production of hydrogen, important materials
questions must be answered.
At the time of the publication of the paper, 86% of the world's energy demands were
satisfied with fossil fuels [3]. The use of hydrogen is expanding as technology permits. In the past, hydrogen has been used extensively in the refining of petroleum and
petroleum products. Hydrogen is also used to produce fertilizers and other chemicals,
such as anhydrous ammonia [3]. Automobile manufacturers have made far-reaching
initiatives to produce hydrogen fuel cell-driven automobiles, as these automobiles can
meet performance demands and produce merely water as a byproduct.
Hydrogen offers excellent benefits when compared to traditional fuels. Hydrogen is
13
clean burning; the combustion of hydrogen results in water. The burning of petroleum
and natural gas release CO2, CO, NOT, and other undesirable greenhouse gases. In
additional to all the greenhouse gases produced from oil and natural gas, coal also
releases dangerous sulfur compounds that must be trapped and converted. Coal is
also not as convenient as the aforementioned energy sources for applications such as
transportation.
Liquid hydrogen also has a high gravimetric energy density of 70.8M-J , whereas gasoline only has a energy density of 44M [4]. The energy density of coal and alternative
fuels, such as ethanol, are about half that of gasoline. An advantage in gravimetric
energy density has great advantages in automobiles, as the fuel weight can be reduced.
Unfortunately, hydrogen is most often available in molecules where it is combined
with other elements, such as water (H 20) or methane (CH4) [4]. Since hydrogen (H2 )
is not widely naturally available, it must be produced from hydrogen containing compounds and existing fuel sources, such as natural gas. Because of this, hydrogen is not
an energy resource, but rather an energy carrier [4]. Since hydrogen is a derivative
of an existing fuel, the energy stored in the hydrogen is less than the energy of the
original fuel and the losses in energy can be directly attributed to inefficiencies in the
production process and storage. The efficiency of hydrogen production ranges from
20% to 80% depending on the method of production [4].
Approximately 95% of the hydrogen demands in the United States are met through
a process called steam methane reforming [4]. Steam methane reforming is a very
efficient process, as methane reforming plants typically operate around 80% efficiency.
However, the major drawback to the process is the process requires the combustion
of natural gas, which is a useful fuel itself and releases greenhouse gases. The steam
methane reforming alone required to meet the hydrogen needs of the United States
contributed to the release of over 74 million tons of carbon dioxide in the last year.
Clearly, finding a means to produce hydrogen efficiently without the negative effects
14
on the environment would be greatly beneficial.
1.1.1
Directions in Hydrogen Production
Hydrogen has been widely hailed as the fuel of the future. However, important decisions must be made in terms of utilization, storage, and production. Many of the
problems have to do with scaling, as demands today are generally for industry. If the
transportation industry does indeed one day rely on hydrogen, the means of production will have no choice but to change.
The necessity to use less refined energy, namely heat rather than fossil fuels or electricity, to produce hydrogen is clear. In terms of efficiency, natural gas and coal
should be used for the production of electricity, and electricity should be spent on the
grid, not to generate hydrogen. Nuclear reactors are non-fossil fuel burning sources
that can produce hydrogen at high efficiencies.
Environmental concerns must also be answered. Dozens of nations have agreed to the
Kyoto protocol, under which carbon dioxide emissions must be reduced on average
5.2% in respect to 1990 levels. One way to achieve this goal is for the transportation
industry to switch from hydrocarbon fuel to hydrogen. The mass adoption of hydrogen as a transportation fuel requires nations to find alternatives to the traditional
method, steam methane reforming.
1.1.2
The Nuclear-Hydrogen Initiative
Hydrogen production may be able to take a great leap with the introduction of Generation IV nuclear technology. Generation IV nuclear reactors meet requisite temperatures for thermochemical methods of hydrogen production, which can produce
hydrogen at efficiencies that have not been approached today without the release of
15
large amounts of CO 2.
Before such methods of hydrogen production are realistic, materials must be developed that can withstand both the temperatures required and the corrosive conditions
present in processes used for production. The development of materials that serve
both as excellent catalysts and can avoid catalytic breakdown under the thermochemical process will not only revolutionize hydrogen production, but it will also aid
hydrogen related products and the field of alternative energy.
1.2
Methods of Production
Many processes have been developed for hydrogen production. The processes vary in
terms of efficiency and reactants. It is important to understand the evolution of these
processes and the different reactions involved. The proper development of materials
for the Sulfur-Iodine Cycle can be aided by understanding the challenges introduced
by thermochemical cycles relative to traditional methods of production.
1.2.1
Steam Methane Reforming
Steam methane reforming involves the combustion of methane to provide heat for the
production reaction, which is endothermic. The first step is a reforming reaction, in
which steam and methane are heated to produce hydrogen and carbon monoxide, as
represented in Equation 1.1. The second step is known as a water gas shift, which is
shown in Equation 1.2; the carbon monoxide produced in the reforming and steam
are reacted over a catalyst to form carbon dioxide and hydrogen. Equation 1.3 shows
the combined reaction.
CH4 + H2 0 16
CO + 3H2
(1.1)
CO + H2 0
CO2
CH 4 + 2H 2 0
H2
) CO2 + 4H 2
(1.2)
(1.3)
Steam methane reforming is a mature technology that provides good efficiency in
hydrogen production. The efficiency of the reaction can reach levels as high as 80%
[4]. However, the drawback is that large amounts of CO2 are generated [5].
1.2.2
Electrolysis
The electrolysis process involves the passing of a current through water molecules to
generate oxygen and hydrogen. The disadvantage of electrolysis is the requirement
of electricity. Since electricity must first be produced, the efficiency of the process
drops dramatically. A typical power plant today operates at around 40% efficiency
[6]. Given that the process of electrolysis is at most 70% efficient, the overall efficiency
of the process is approximately 25-30% [5].
Electrolysis consists of two separate half reactions - a cathodic and an anodic one.
The cathodic half reaction consists of the combination of hydrogen ions and electrons
to produce hydrogen gas, as represented in Equation 1.4. The anodic half reaction
involves the conversion of water into oxygen gas, hydrogen ions, and free electrons,
as shown in Equation 1.5. Equation 1.6 shows the combined reaction, where water is
used to produce oxygen and hydrogen.
2H+ + 2eH2 0 -H20
1
2
, H2
02 + 2H + 2e1
2
02 + H2
0,
(1.4)
(1.5)
(1.6)
Figure 1-1 shows a theoretical layout of an electrolysis cell. The main efficiency and
17
cost drawbacks of traditional electrolysis are due to the high consumption of electricity.
®
(O
H20 -1/202
i
-- 0
+ 2H++ 2e'
2H++ 2e--
_
Anoae
.
H2
r
Latnoae
Figure 1-1: Electrolysis Cycle
There are also environmental issues to be considered depending on the source of electricity for electrolysis. If a power plant operating on traditional energy sources, such
as coal or natural gas, is used to produce the electricity for electrolysis, there will
be CO2 emissions. Since the electricity demands are so great, there are significant
environmental tradeoffs.
1.2.3
High Temperature Steam Electrolysis
High temperature steam electrolysis (HTSE) trades some of the energy used for electricity in traditional electrolysis for heat. The addition of heat reduces the electricity
demands of the process, and the total energy necessary is less than the energy for low
temperature electrolysis.
The only inefficiencies in heat transfers are due to limitations in heat exchanger technology, which are minor in respect to turbine and compressor technologies. Through
18
high temperature steam electrolysis, efficienciesof 35-40% can be achieved [5].
High temperature steam electrolysis is also well-suited for Generation IV nuclear reactors, since they can use combined electric and steam cycles at high temperatures
for the HTSE.
1.2.4
Sulfur-Iodine Cycle
The Sulfur-Iodine cycle was invented by General Atomics in the 1970s. The cycle
provided great benefits compared to earlier methods of hydrogen production. With
current technology, the cycle produces hydrogen at an efficiency of 47%, and by combining hydrogen and electricity production, efficiencies of up 52% can be achieved [4].
Like electrolysis, the only inputs necessary for hydrogen production using the SulfurIodine cycle are water and heat. A molecule of water is combined with a molecule
of iodine, sulfur dioxide, and another molecule of water. At 120°C, the reactants are
converted to a molecule of sulfuric acid and two molecules of hydrogen iodide. The
cycle diverges from this point.
The sulfuric acid is passed over a catalyst at a temperature greater than 830°C, which
causes the sulfuric acid to decompose and leads to half a molecule of oxygen, and a
molecule each of sulfuric acid and water. The sulfuric acid and water are returned as
the reactants used initially. The oxygen is a product of the reaction.
Hydrogen iodide from earlier in the cycle is heated to form hydrogen and iodine. The
iodine is used as a reactant in the first stage of the cycle discussed, and the hydrogen
is the desired product of the reaction. Thus the only input into the cycle is water
and heat, and the products of the reaction are oxygen and hydrogen.
19
The Sulfur-Iodine cycle can be summarized with the following chemical reactions:
H2 SO4
1
-
SO2 + H2 0 + -02
2
12+ SO2 + 2H2 0
2HI
H20
) 2HI + H 2 SO4
(1.7)
(1.8)
(1.9)
) 12 + H2
1
(1.10)
) H2 + O02
2
The cycle is displayed in Figure 1-2.
1/202
Catalyst, 830°C
_-
H2S04
1/202+ S02 + H20
120°C
H2S04 + 2HI
2HI
12+
S02+ 2H204 .... H20
320°C
12 + H2
H2
Figure 1-2: Sulfur-Iodine Cycle
Clearly, the Sulfur-Iodine cycle is a desirable cycle, both in terms of environmental
friendliness and high efficiencies. However, there are still many questions to be answered about production.
20
The conversion of sulfuric acid to sulfur dioxide, Equation 1.7 is a reaction that poses
great challenges within the Sulfur-Iodine Cycle. The reaction can be broken down
into two individual reactions.
H2 SO4
-
S0 3 -
SO3 + H 2 0
1
SO2 + -02
2
(1.11)
(1.12)
High temperatures and an excellent catalyst are required to facilitate this reaction,
and the development and testing of materials of this reaction are the focus of this
research.
1.2.5
Summary of Methods
The aforementioned cycles highlight one major development in the progression of
methods of hydrogen production.
The source of energy for hydrogen production
has advanced towards lower quality energy sources, such that high quality, valuable
sources do not have to be used in the production of an energy carrier like hydrogen.
Natural gas and electricity are both high quality sources of energy. In terms of trans-
port and convenience,they are both outstanding.
Clearly, it would be preferable to produce hydrogen from other sources, and the desire
for this is seen in high temperature steam electrolysis, where a great deal of electricity
is substituted for heat, which is a low quality energy source. Even further along in
the progression are thermochemical methods, which rely entirely on heat as an energy
source. Using heat to produce hydrogen is ideal, since distributing and utilizing the
heat directly for commercial applications is difficult, while converting it to electricity
requires losses in energy.
21
1.3
Difficulties in Production
At the operating conditions of the more advanced processes, such as HTSE or the
Sulfur-Iodine cycle, material, structural, and thermal questions must be answered.
The step in which sulfur dioxide is produced from sulfuric acid in the Sulfur-Iodine
cycle is of great interest. The production of sulfur dioxide requires passing sulfur
trioxide over a catalyst at temperatures greater than 8000 C. Along with the corrosive
conditions, large pressures and excellent catalytic ability lead to additional requirements. The supply of heat to the Sulfur-Iodine Cycle will require the presence of an
interface between the energy or heat producer and the process itself. This interface
will likely take the form of a heat exchanger. The required operating temperature of
the heat exchanger will be at least 8000 C. The energy transferred will then be used
for the process-which will require a catalyst. If a heat exchanger that could serve
both heat exchange and catalytic functions could be developed, the process could be
simplified considerably. The work in this thesis addresses the development of such a
heat exchanger.
1.3.1
Thermal Considerations
The efficiency of the Sulfur-Iodine Cycle is highly dependent on temperature [2]. For
the cycle to be desirable relative to electrolysis, the minimum operating temperature
of the cycle should be approximately 800C. Current plans for Generation IV reactors
should be able to produce efficiencies in the 50% range, which is already excellent.
Future advances in reactor technology may be able to accommodate even higher temperatures, and the cycle efficiency could eventually approach 60%.
Concerns arise because of the high operating temperatures.
The corrosive environ-
ment created due to the presence of sulfuric acid and other decomposition products is
only augmented by the temperature. The catalytic material must be able to maintain
22
E %
70%
640%
g
10%
20%
600
700
800
900
Temperature (deg. C')
1000
Figure 1-3: Temperature Dependence on Efficiency of the Sulfur-Iodine Cycle
(Source: [2])
catalytic ability in the long run.
1.3.2 Structural Considerations
Because of the application in a nuclear reactor and the exposure to the highest temperatures exiting the reactor, the heat exchanger almost will certainly ultimately be
exposed to gas at the maximum operating temperature and pressure of the reactor.
In terms of Generation IV reactors, this pressure will likely be in the range of 7 MPa,
so the material must meet pressure code specifications for high pressures.
1.3.3 Past Research
Studies have been performed on the catalyst stability of platinized A12 0 3 , TIO02, and
ZrO 2. [7] Platinum concentrations of 0.1% by weight and 1% by weight were used.
The platinum-coated ZrO2 resulted in the greatest SO 2 production, however, regardless of the platinum concentration, the materials suffered extreme deterioration [7].
The sulfuric acid penetrated the platinum layer and lead to catalyst deactivation and
23
break down [7].
1.4
Material Selection
The catalytic material will be used as a heat exchanger material in the future for the
Sulfur-Iodine cycle. Material properties that result in a good heat exchanger are also
desirable.
Platinum is inert so it is not susceptible to the effects of corrosion. Due to cost, thermal, and structural considerations, pure platinum is not an optimal choice. Platinum
is often used as a coating for other alloys, which drastically reduces the amount of
platinum that is required [7].
However, even coated materials have shown poor stability in the presence of high
temperature sulfuric acid [7]. At temperatures of greater than 800° , there are kinetics that result in the exposure of the underlying alloys.
Because the underlying material contains no platinum, it has no defense to the high
temperature sulfuric acid. A reasonable consideration would be to add platinum to
the underlying material. Adding a large amount of platinum is not desirable for the
reasons discussed earlier, but the platinum could be alloyed with other metals.
Incoloy 617 and 800H are superalloys with excellent oxidation resistance at high
temperatures. The oxidation resistance of both of the materials is enhanced by the
addition of aluminum. These materials are widely used in turbines, superheaters, and
furnaces, so in terms of a corrosion and temperature perspective, they are excellent.
In this thesis alloys with the base composition of Incoloy 617 and 800H have been
alloyed with small amounts of platinum. The initial evaluation of the performance of
24
these materials as catalysts is the subject of this thesis.
1.5 Determination of Catalytic Ability
The most straightforward means to characterize the catalytic ability of the alloys
developed is to utilize the alloys in the Sulfur-Iodine cycle. Sulfuric acid could be
heated up to reactor conditions and passed over the catalyst. The resulting sulfur
dioxide produced can be measured, and the catalytic ability of the material can be
determined.
The testing of the material in this manner requires the construction of a loop that is
capable of reaching reaction conditions. The catalytic ability and the alloy stability
in the environment can then be directly determined.
25
26
Chapter 2
Methodology
2.1 System Requirements
The main requirement of the system is the ability to provide reaction conditions
necessary for the catalysis of sulfuric acid to sulfur dioxide. The first step is the
decomposition of sulfuric acid into sulfur trioxide and water, which is represented by
Equation 1.11 and presented once again below.
H2 SO4
-
SO 3 + H 2 0
Prior to this reaction, the vaporization of the liquid sulfuric acid, the decomposition
into sulfur trioxide, and temperatures of 340°C are necessary. Precautions should
be taken in limiting the mass flow rates, as the high temperatures require the quick
vaporization of the sulfuric acid.
The decomposition must be accomplished sufficiently prior to the interaction with
the catalyst, as appreciable reduction of the sulfur trioxide represented by Equation
1.12 requires temperatures at or above 800°C, which is presented once again below.
SO 3
,SO
27
1
2
+ 102
2
Once the reduction of sulfur trioxide is catalyzed, the sulfur dioxide concentration of
the exit stream must be measured. Because of the harmful nature of sulfur trioxide,
any remaining sulfur trioxide should be converted back to sulfuric acid by cooling in
the presence of water. The sulfur dioxide and oxygen produced should remain in the
system and then can be quantified.
2.2
Loop Construction
Figures (2-1)-(2-8) show the system and its components. Sulfuric acid (96% by weight
in water) is pumped through teflon tubing from an Erlenmeyer Flask at ambient pressure using an Ismatech micrometering pump. A micrometering pump is used to slowly
pump the sulfuric acid in order to limit the overall flow rate to enhance residence time
and limit the magnitude of the pressure spikes due to the expansion from fluid to vapor. At the entry to the peristaltic metering pump, the telfon tubing is connected
to viton tubing, which can withstand over 100 hours of the compression due to the
peristaltic action and the corrosive nature of the sulfuric acid. The sulfuric acid reenters a network of teflon tubing and passes through a check valve before entering the
sample chamber and the furnace as to ensure no back-flow from the vaporization of
the sulfuric acid and subsequent pressure increase.
Figure 2-1: Ismatech Micrometering Pump
28
The sulfuric acid decomposes into water and sulfur trioxide at 3400 C in the furnace.
The gaseous sulfur trioxide continues to heat up until it reaches 9000C. Afterwards,
the sulfur trioxide passes over the catalyst, and the reaction to form sulfur dioxide
may be catalyzed.
Sulfur dioxide, sulfur trioxide, water, and oxygen then leave the reaction vessel and
are cooled. Any uncatalyzed sulfur trioxide then joins with a molecule of water to
reform sulfuric acid once the temperature drops below the decomposition temperature. Since water is in excess, the remaining sulfur trioxide should combine with it,
condense, and collect in an Erlenmeyer flask at the outlet. The excess water also
condenses and collects in the outlet flask. The product stream should only contain
sulfur dioxide and oxygen at this point. Nevertheless, the stream is passed through a
second collection Erlenmeyer Flask in case there is still sulfur trioxide and water in
the stream. A check valve separates the first and second collection Erlenmeyer flasks
to prevent back-flow.
Inside the second Erlenmeyer flask, nitrogen is added to serve as a reference such
that the sulfur dioxide concentration in the exit stream can be quantified in the gas
chromatograph. The product gas stream then passes through another check valve,
and finally enters the gas chromatograph.
Figure 2-2 shows the Piping and Instrumentation Diagram (P&ID) of the system.
29
Micrometering
Pump
S03+SO2+
I
Figure 2-2: Piping and Instrumentation Diagram
2.3 Reaction Vessel Design
The sulfuric acid must properly heat up such that it boils, decomposes, and reaches
a temperature of greater than 850° prior to reaching the catalyst. The necessary preheating makes the addition of a preheat loop to the reaction vessel favorable, as such
a loop will allow the reactants to reach the proper temperature, along with isolating
the violent expansions the liquid sulfuric acid to the gaseous sulfur trioxide and water
vapor from the catalytic reaction. Figure 2-3 shows the completed reaction chamber.
30
I
-
b'!
--
11,.-..
.
t 11
i
I
\
I
W
-,
' J
1_1
g_.'
E-4
I
---
I
;....
1- r -- -- ---- -- -- -- ---- -- ---- -- -- -- ---- r-I
I .l11
i
I
i.....
=
I
I
17
F
II
-i4
U i.1i.I
!
I
..-I
_ r.c ..
-
Figure 2-3: Reaction Chamber, Units in Inches
Additionally, the temperature of the gas at the outlet of the reaction vessel was required to stay within the bounds of approximately 100° . Due to the 190°C melting
temperature of the following teflon tubing and fittings, maintaining a subsequent
system temperature of significantly less than melting temperature was necessary to
ensure no compromise of the integrity of the tubing.
The furnace placed a limit of 1.5" on the entire diameter of the reaction vessel. The
furnace had a length of 14.5" with a heating length of 11", so there was a constraint
on the maximum heated area possible.
The first consideration was to maximize the length over which the sulfuric acid is
heated. The entire 11" heated region was used for a
"
OD preheating tube. The
end of the tube looped around in a semicircle and then expanded into a
"
OD.
To ensure that the surrounding teflon tubing remained well below the melting temperature, the reaction vessel was extended 6 inches from the inlet of the tubing into
the reaction vessel. Factoring in the insulation of the system, this distance was significantly more than necessary to maintain near ambient temperatures.
Additional calculations were necessary for the outlet end to ensure that the mass flow
of the product stream could not cause enough of a heat transfer such that the tubing
31
could eventually reach the melting temperature.
Complicating the calculation was
the condensation of components in the product stream. Equation 2.1 represents the
minimum and averaged heat transfer coefficient for the product stream.
4
4-(g h f pP( - P)kS3¼
hav= 3 hL= 3-(
)4
4pLATT
(2.1)
ha,,= 0.943 (hfgP(pT)k
ptLAT
The convective heat transfer coefficient can be found by simply inputting all the
constant terms. The mean of the expected minimum and maximum constants was
used as an approximation for many of the values.
Term
Value
g
9.81 m
hfg,min
hfg,max
2256k
hfg
11283
Pmin
322 kg
Pmaz
0
1797.3
k3
P
1059.7
Pv,min
0.59817m
Pv,maz
322-
Pv
161.299
kf,min
kf,max
kf
0.42504
0.67909
0.55207.
k
A/min
5.2069E-05
/max
2.8174E-04
c
1.6690E-04 me
P
N7ec
AT
800°C
R
0.18in
Table 2.1: Fluid Constants
T -Tb,out
T - Tb,i n =
32
e- 2Rh L
e
cp
293 - 373
293-
1223
= e
-"R
rRo.943(
gh P(P-Pv)k 3'
-_ ¼"4__'_'
_)
Li
mhcp
m = .029
cp = 24.02
kgK
L = 7.018cm
The calculation reveals that only a few inches are necessary to cool the fluid to temperatures comfortably within the operating range of the teflon. For precautionary
purposes, the outlet length was expanded to 10 inches in the case operating flows
were increased.
The reaction chamber was made out of quartz, which has a melting temperature of
1600°C, which is comfortably greater than the maximum system temperatures.
2.4
Hardware Interfaces
One 0-5 kg range Mettler Toledo balance was placed below each of the inlet and outlet
Erlenmeyer flasks, as shown in Figure 2-4. The sides of the scales were lined with
cardboard to limit the variance in the scale readout due to thermal convection and
exhaust fan-induced flows within the hood.
33
Figure 2-4: Mettler Toledo Scale
One 1179A mass flow controller and one 179A mass flow meter from MKS Instruments
were used, as shown in Figure 2-5. The mass flow controller was placed immediately
after the bottle of nitrogen reference gas in order to measure and control the flow
for later quantification of the sulfur dioxide. The mass flow meter was placed at the
exit of the second collection bottle. It was used to measure the total flow entering
the chromatograph. The second mass flow meter aided the detection of appreciable
changes in mass flow and the tracking of gas flow in respect to the known flow exiting
the mass flow controller prior to sampling by the chromatograph.
Figure 2-5: 179A Flow Meter
34
The chromatograph was a 6890N model made by Agilent Technologies, as shown in
Figure 2-6. The chromatograph required three separate gas lines. The first line was
a compressed air line for the air-actuated valves. A dryer, as shown in Figure 2-7 was
connected to the input air line to prevent condensation of liquid inside the actuators.
The two other lines were reference and carrier gas lines. Helium gas at 50 psi was
connected to both of these lines. Because of the use of helium as a referemce, any
helium passing through the chromatograph would not be detected.
Figure 2-6: 6890N Agilent Technologies Gas Chromatograph
Figure 2-7: Dryer Connected to Compressed Air
The final catalytic loop is shown in Figure 2-8.
2.5 Hardware Settings
The two electronic balances were manually set to output in floating point format and
be accessible to remote transfers from the settings menu. A Keyspan 4 port serial
interface was used to link the scales to the data acquisition system. The data is transfer is begun with a command of a "SIR" and a carriage return and line feed, which
repeated returns the data at particular time intervals. The scalewas instructed to not
35
Figure 2-8: Catalytic Loop
wait until measurements stabilized, as the masses in the inlet and collection bottles
were expected to be constantly changing. A MSComm interface in Visual Basic was
used to interpret the system interrupts.
The mass flow controller was calibrated to 20,000 sccm of nitogren. The mass flow
meter was calibrated to 30,000 sccm of helium, but the scale factor led to an equivalent range of 0-20,000 sccm of nitrogen. The two mass flow meters were connected
to an MKS display, and a 25-pin D-connector was used to read the signals from the
mass flow meters. A HP3852 Data Acquisition Unit and a voltmeter module were
used to read the voltages from the mass flow meter and controller. The voltages were
0-5V full scale, and the conversion between voltages and the mass flow readout was
4000 minV'
cmin
36
Figure 2-9: Flow Readout Panel
Figure 2-10: HP Data Acquisition System
The system was set to call for temperature readings, mass readings, and mass flow
readings every second. Data was exported to a text file. MATLAB was used for data
interpretation.
2.6 Material Preparation
Each of the samples, 2, 5, 15, and 30 % Pt in 600 or 800H, were rolled to .0023". The
samples were annealed in hydrogen-rich environment, and they were finished with a
cold roll.
The samples were lightly finished to remove surface oxides. Hand grinding was performed with 600 grit silicon carbide paper in a wet environment. Grinding continued
until all visible oxide was removed.
The samples were then cut to strips. Creases were folded into the strips to facilitate
packing into the reaction vessel.
2.7
Material Installation
The tapered section of the outlet of the reaction vessels were cut at the point when
the reaction the
vesselractio
expanded
to o~4". Samples were loosely packed into a 3.5" segment
vessl expnded
37
of reaction vessel to ensure a large amount of catalytic surface relative to the reaction
vessel volume in he region. The sample vessel was then resealed.
Figure 2-11: Catalyst Sample in Reaction Vessel
2.8
Chromatograph Setup
The chromatograph was equipped with a packed inlet flow source, a 6-port gas sampling valve (valve 1), and a 6-port column isolation valve (valve 2). A Porapak
N
column was installed as the first column, onto which valve 1 enables flow. A Molecular
Sieve Column was installed on as the second column, or isolation column. Flow onto
the isolation column is normally open when valve 2 is closed. Turning on the second
valve isolates column two from the sampling mechanism, or thermal conductivity detector (TCD). Returning the second valve to the off position allows for sampling of
the gases that were isolated. Figure 2-12 shows a layout of the valves.
38
Figure 2-12: Valve Layout
The gas stream was expected to contain nitrogen, oxygen, and sulfur dioxide as its
main constituents. When the first valve is enabled, the oxygen and nitrogen pass immediately over the Porapak N column and into the Molecular Sieve Column. By then
turning on valve 2, the nitrogen and oxygen are isolated from detection. During this
period, the sulfur dioxide flows through the Porapak N column and into the detector.
Once the sulfur dioxide has been detected, valve 2 is turned off. The oxygen and
nitrogen elute into the TCD and are detected. Afterwards, the temperature of the
gas chromatograph was ramped up to allow any remaining gas in the sample line to
bake out.
The detection of the gases is time based, so the time of isolation and elution are
absolutely necessary in proper measurement. At the beginning of the measurement,
39
valve 1 was turned on. The time allowed for the oxygen and nitrogen elution was
set at 0.8 minutes as recommended by Agilent Technologies. The second valve was
turned not turned off until 3 minutes into the detection, which allowed ample time
for loading the sample valve and elution of the sulfur dioxide into the detector. After
the second valve was turned off, the oxygen and nitrogen would elute into the detector.
2.8.1
Chromatograph Calibration
The chromatograph was calibrated with nitrogen, oxygen, and sulfur dioxide. Initially, nitrogen was pumped into the system. The nitrogen calibration curve is displayed in Figure 2-13. The analysis of the nitrogen sample yielded only one significant
peak, as expected.
25uV. ·
1
2
3
4
5
6
7
min
Figure 2-13: Nitrogen Calibration Curve
Integrating the peaks of the chromatograph analysis produced the results shown in
Table 2.2.
Peak
Time
Area
Height
Width
Area % Symmetry
1
2
3
4
5
6
9.968E-2
0.814
3.019
3.809
3.957
4.74
35.2
222.6
50.3
56.6
76460.1
13.5
8.8
19.5
28.9
12.5
11023.1
1.9
0.0626
0.137
0.0303
0.0707
0.1049
0.1123
0.046
0.290
0.065
0.074
99.508
0.018
0.975
0.245
0.7
0.902
0.268
0.556
Table 2.2: Integration Results of Peaks of Nitrogen Calibration
40
The nitrogen displayed one main peak at about 3.9 seconds. The signal resulting in
peak 2 and peak 3 were from the firing of valves. The signal at peak 4 was dependent
on the amount of air in the sample stream prior to sampling. Depending on the air
initially in the sampling volume, a second peak at about 3.8 seconds also appeared.
Because of the concentration of gases in air, the only feasible source of the peak was
oxygen.
A gas mixture of 5% sulfur dioxide in helium was used for calibration of the sulfur
dioxide peak. Since the reference and carrier gas used by the chromatograph system
was helium, the helium inside the gas mixture was undetected. The results of the
analysis are shown in Figure 2-14.
25iuV
140
130
120
110
100
90
80
70
60
50
1
2
3
4
5
6
7
min
Figure 2-14: Sulfur Dioxide Calibration Curve
There were relatively few integration events of concern for the sulfur dioxide calibra-
tion. The results of the integration are shown in Table 2.3.
Peak
Time
Area
Height
Width
1
1.798
7
1.8
0.0587
1.724
0.747
2
2.035 388.2
88.8
0.067
95.586
0.715
3
2.745
1.7
0.0987
2.689
0.702
10.9
Area % Symmetry
Table 2.3: Integration Results of Peaks of Sulfur Dioxide Calibration
Table 2.4 shows the approximate times at which the gas peaks were expected to appear based on settings used in calibration. The times differed slightly in respect to
41
the original calibration tests because of the presence of the gases together as a single
sample.
Time(min)
Detection Event
2.0
SO2
3.7
02
3.9
N2
Table 2.4: Calibration Values
2.8.2
Valve Timing
After measuring approximate peak times, the valve timing was set as described in
Table 2.5. The table shows the valve switching, and the expected location of the gases
at each significant change in the system settings. Since the oxygen and nitrogen are
to be isolated in the Molecular Sieve column, they are detected after the sulfur dioxide.
Time(min)
0.00
Valve 1 State
0.01
ON
ON
ON
ON
ON
ON
0.80
2.0
3.00
3.7
3.9
7.00
I Valve
2 State
-
_
OFF
OFF
OFF
Porapak N
Gases
NONE
SO2
SO 2
ON
ON
Molecular Sieve
Gases
NONE
N2 & 02
N2 & 02
N2 & 02
N2 & 02
NONE
OFF
NONE
OFF
NONE
N2
NONE
NONE
OFF
NONE
NONE
OFF
OFF
Table 2.5: Valve Timing and Detection Events
.
Detector
Gases
NONE
NONE
NONE
_
SO 2
NONE
02
N2
NONE
The operation of the cycle from 0.00 minutes to 7.00 minutes was performed at 80°C.
After the 7 minute mark, the temperature was ramped up to 120°C to bake out any
trapped gases remaining in the columns. The baking was performed to clean the
columns for subsequent runs.
42
Chapter 3
Results
3.1 Mass Flow Rate Analysis
A reaction vessel packed with the rolled and cut samples of 800H + 5% platinum
was examined. Sulfuric acid was pumped through the loop until it reached the entry
point of the reaction vessel. The furnace was then turned on and heated until reaching the final steady state temperature. As the temperature was raised, the sulfuric
acid was pumped and decomposed through the reaction chamber until the desired
temperatures were reached. Upon reaching the steady state temperature, the system
was allowed to run at steady state for 0.69 hours.
The mass of the inlet vessel containing the sulfuric acid and the outlet collection vessel were monitored throughout the course of the experiment. Two separate analyses
of the product gas stream were performed during the steady state period. Mass flow
controllers were used to control and monitor the flow of the nitrogen used as a reference and carrier gas for the chromatograph analysis. The temperature and mass flow
data are represented in Figure 3-1. Following the data collection, the reaction vessel
was slowly cooled. The micrometering pump remained on and all data collection
continued for another 3 hours.
43
Temperature
---A
1 UU
0
80
Qa
0
60
0.
E
40
a,
I-
20
2000
4000
6000
8000
10000 12000 14000 16000 18000
Time(s)
Scale Mass Balance
600
400
l
Mass In
200
m
Mass Out
:
-200
'
0
- - I2000
- -1 4000
IIIIII
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
6000
8000 10000 12000
Time(s)
14000
16000 18000
Figure 3-1: Temperature and Mass Flow Data
Data points for temperature, mass in the inlet and collection vessels, and mass flows
were recorded every second for a period of 4.5 hours. Steady state temperature was
reached once the operating temperature was within 5°C of the desired operating temperature of 903°C.
3.1.1
Mass Flow Rates at Steady State Temperature
The system reached a steady state temperature of 903±1.9C at 0.97 hours into the
experiment. Prior to reaching steady state, there were significant pressure gradients
in the system due to the boiling and decomposition of the sulfuric acid and the pressurization of the air in the loop. The fluctuations are evident by the initial mass
decrease in the outlet vessel in Figure 3-1. Once steady state was reached, the mass
44
flows linearized. The temperature and the mass during the steady state period are
shown in Figure 3-2.
Steady State Temperature
910
---
....
rrr
a,
· r
2a 905
0.
r
r
r
··
I
r
· · ·
rr
E
I- 900
E
2
nne
MU-h
I
i
3500
4000
-
4500
--
___-
5000
5500
6000
5500
6000
Time(s)
Steady State Scale Mass Balance
rin
ouu
,
400
*
u,
l,0
Wb 200
co
*
Mass In, -0.01159t+0.18391
Mass Out, 0.00927t+437.65
O
-
n1'
_3500
l
4000
4500
5000
Time(s)
Figure 3-2: Steady State Temperature and Mass Flow Data
Least squares analysis revealed the mass within the vessel containing the sulfuric acid
was -0.01159t + 0.18391 g during the period of steady state temperature. The linear
least squares model of the mass of the outlet collection vessel was 0.00927t + 437.65 g.
The corresponding inlet and outlet mass flow rates were then -0.01159
i
and 0.00927
respectively. The difference in the mass flow rate of the inlet and outlet collection
vessels, 0.0023
can be accounted for by the ideal reduction of SO3 . According to
equation 1.12, half a mole of oxygen is produced for every mole of SO2 produced in
the reduction of sulfur trioxide.
45
SO3
-
SO2 + 202
Correspondingly, for every 16 grams of oxygen produced, there are 64 grams of SO2
produced. Thus 20% of the product mass can be attributed to
02
production, while
the other 80% is attributed to SO2 production. Under ideal conditions where no corrosion products contribute to the difference in flow rate, the entire difference is due
to the product mass. Knowing this, the ideal mass flow rate of the SO2 produced is
.0018 a.
3.1.2
Mass Flow Rates as a Function of Temperature
The period during which the furnace was turned off and the furnace temperature
decreased provided interesting time based data. Since the data of the temperature as
a function of time is present, the optimal catalytic activity of the system as a function
of time can also be calculated. The temperature and mass curves are shown in Figure
3-3.
46
Temperature
1
o
q.,
e
a)
aE
a)
Q.
00
Time(s)
Scale Mass Balance
600
400
*
U3
{33
U3
C,
Cu
200
Mass In, -0.01159t+0.18391
Mass Out, 439.95392e
co
1 9 19 1 E - 5
1
0
o~~~~~~~~~~~~
-200
60)00
6500
7000
7500
8000
8500
9000
Time(s)
Figure 3-3: Mass Flow Data with Dynamic Temperature
The mass into the system occurs at the same rate as for the steady state period,
because the micrometering pump operates independently of the rest of the system.
The mass collected in the outlet flask accumulated at a dynamic rate because the
temperature of the system was changing, and thus it was expected that the sulfur
dioxide production would change as the temperature changed.
Performing the same calculations as in Section 3.1.1 and then curve fits, Figure 3-4 is
found. The production at the steady state temperature varied slightly from the earlier
calculations, as the original steady state temperature had a variance of ±1.9°C.
47
-1
I1.00
X 10
- 3
Unaccounted Mass Rate as a Function of Temperature
1.6
1.55
1.5
Z
0)
IV 1.45
a:
1.4
2
1.35
1.3
1
r
500
550
600
650
700
750
Temperature
800
850
900
950
eC)
Figure 3-4: Unaccounted Mass Flow Rate in Respect to Temperature
The plot displays the mass flow rate difference as a function of temperature.
Note
that the mass flow rate at the maximum temperature of the experiment is slightly less
than 1.8x 10-3
predicted early due to the aforementioned variance during the steady
state. The maximum possible production decreases as the temperature decreases.
3.2
Gas Chromatograph Analysis
The insufficiencies in the calculation of the SO2 production using the difference in
mass flow rates lie in the formation of corrosion products. The general composition
of Incoloy 800H is described in Table 3.1.
Because of expected oxide and sulfide formation, the mass flow rate difference is not
entirely due to the formation of the desired products. The idealized calculations of
48
Element
Relative Amount (%)
Nickel
Chromium
Iron
Carbon
Aluminum
Titanium
30.0-35.0
19.0-23.0
39.5 minimum
0.05 - 0.10
0.15 - 0.60
0.15 - 0.60
Table 3.1: Composition of Incoloy 800H
SO2 production can have significant error. A more accurate means of confirming and
quantifying the SO2 produced is to directly detect and measure the SO2 using a gas
chromatograph.
3.2.1
Gas Chromatograph Analysis During Steady State Tem-
perature
The Agilent Technologies G1540N 6890 gas chromatograph was used to detect the
SO2 catalyzed. Five minutes after the system loop reached steady state temperature,
a 0.5 cm3 sample injection was made into the gas chromatograph. The time-based
peaks detected in the injection are shown in Figure 3-5.
25uV
12000
10000
8000
.ooi
6000
4000
2000
2
0
1
2
3
3
7
4
3
4
i
6
7
min
Figure 3-5: Chromatograph Analysis at 5 Minutes of Steady State Period
Figure 3-5 shows the expected nitrogen peak at 3.88 seconds. There is a large peak
49
at approximately 0.8s that is not a result of expected gases. The peak is due to
a false positive in the signal detection due to the firing of valve 2. Although that
false peak is large relative to the other peaks in the sample, it does not have bearing
on the integrity of the data since the other signals are measured relative to each other.
Using the enhanced autointegration feature within Chemstation, the SO2 peaking at
2.04 seconds is detected. The SO2 peak is evident upon magnification of the 2.04
second point. There is a clear peak in the signal, which is represented by the blue
line in Figure 3-6, over the background signal, which is represented in purple.
-)c.%1
Lvuv
140
120
2
100-
~----.-
80-
60-
.
.
.
.
.
.
.
.
.
.
.
.
.
*_------~_~_~
401.90
1.95
2.00
2.05
2.10
2.15
2.20
2.25
2.30
min
Figure 3-6: SO2 Peak at 5 Minutes of Steady State Period
The enhanced integrator in the Chemstation software was also used to calculate the
peak times and characteristics. The results of the integration are provided in Table
3.2.
Notable in the integrated values is the correlation of the SO2 and N2 peaks with the
original calibrated peaks. The nitrogen peak is significantly larger than the SO2 peak,
but the magnitude difference was expected, as to get appreciable flow through the
gas regulator, a significant pressure was required, and even the minimum levels of the
mass flow controller were on the order of a few hundred the
"--.
These
factorsThe
resulted
nitrogen
many
detected
to
be
orders
greater
than
the
SO2
detected.
relative in
the nitrogen detected to be many orders greater than the SO2 detected. The relative
50
Pea] k Time
1
0.846
2
2.035
3
2.521
4
3.02
5
3.712
6
3.882
7
6.044
Area
Height
Width
Area % Symmetry
.
37762.2
99.6
47
315.4
12059.3
60488.3
5298.5
13647.6
9.1
0.0413
0.1333
0.1083
0.0304
0.0696
0.0948
1.1548
R
5.6
134.3
2624.2
9559.1
53.7
-
-
0.041
.392
0.829
0.915
0.272
0.18
10.390
52.113
0.692
0.185
0.641
32.534
0.086
4.565
Table 3.2: Integration Results of Peaks at 5 Minutes of Steady State Period
_
_
percentage of the nitrogen to the relative percentage of the sulfur dioxide, coupled
with the known flow rate of the nitrogen can be used in Equation 3.1 to find the mass
flow rate of the sulfur dioxide.
*
SO2 = PN2 VN2
so2 Aso 2
%N
AN 2
(3.1)
-ON2 AN2
The density of the nitrogen at standard temperature and 10 psi is 1.9458 kg. The
flowrate recorded is 569±7 -.
The flow rate during the steady state temperature
is shown in Figure 3-7.
51
MassFlow(Gas)
1300
ControlledGas Flow
MeasuredTotalMassFlow
-
1200
1100
-
1000
E 900
-
II
I
A
.1
11
-
800
700
600
500
1
3500
.
4000
,
4500
,
5000
I
"I
I
5500
I
6000
Time(s)
Figure 3-7: Steady State Nitrogen Mass Flow Rate Data
kg cm3 x 0.086 --6415
= 7.034x 10 s
rhso = 1.9458m3 x 569min
52.113 28
Clearly, this value is significantly lower than the maximum production. Possible explanations for the low value found under the current analysis may lie in the presence
of the oxygen in the system. Some of the oxygen in the system is due to the product
formation during the reduction of the sulfur trioxide, however, as mentioned earlier,
the oxygen levels should be lower than the sulfur dioxide levels. An explanation for
the high oxygen levels may be due to air that had been in the system or had been
pumped into the loop.
The oxygen introduced by the presence of air can be easily corrected for, as the oxygen due to the reduction of the sulfur trioxide is simply half the molar amount of the
SO2. The rest of the oxygen can be assumed to be due to the presence of air, and a
corresponding amount of nitrogen due to air can also be found.
52
1
Areao
2 from SO
- Area%so 2 = 0.043
Area%o0 from air = Area%o 2 - Area%o 2 from SO3 = 10.344
Area%N from air = Aea
from air N 2 in air = 1
0
.089476
Area%N2 from reference = Area%N2 - Area%N2 from air = 13.555
Now that the actual relative amount of nitrogen in from the reference gas has been
found, the original calculation for the SO2 mass rate can be recalculated.
rhso2 = 1.9458
m
k
3
cm 3
0.086
x 3 55x
min
13.555
x 569-
64
-49
- = 2.676 x 10 428
s
The corrected value is approximately an order of magnitude lower than the maximum
production.
A second gas sample was analyzed by the chromatograph at the 25 minute mark of
the steady state period. Figure 3-8 shows the time based analysis of the sample. The
valve switching did not create a peak of the same magnitude as earlier, thus there is
only minor peaking at the 0.8 second mark.
53
VA-.1
10000-
8000-
6000-
4000.
2000.
6
1
2
3
,+O- · .
0
-I-c-*-·---·-c--·*--·--·r·------c-1
*
8
5
4
I....
-
-,--·T------c-1------T------,----c--5
4
3
2
mln
6
Figure 3-8: Chromatograph Analysis at 25 Minutes of Steady State Period
The presence of the SO2 peak is once again confirmed upon closer inspection. Peak
2 represented in Figure 3-9 occurs at 2.04 seconds. Because the opening of valve 2
created much less of a signal in this sample, the background baseline is more level
than the earlier run.
25UV
275
250
225'
200
175'
150
125
100
3
-
--
-
75'
snr -
·
1.95
2.00
·
·
-
-
2.05
-
2.10
2.15
min
Figure 3-9: SO2 Peak at 25 Minutes of Steady State Period
Table 3.3 displays the integration results obtained by using the enhanced integrator
feature in the Chemstation software. The earlier analysis used to obtain the mass
flow rates of the SO2 from the comparision of the SO2 and the N2 signals can be used
again for this case.
54
Peak
Time
Area
Height
1
0.811
0.849
2.04
2.525
3.02
3.721
3.878
6.058
12.9
880.9
77.7
21.1
177.7
4064.3
67208.4
3815.8
21.7
112.1
7.6
2.5
109.4
921.1
10436.7
40
2
3
4
5
6
7
8
Width
I
Area % Symmetry
9.8913E-3
0.017
0.568
0.1046
0.1226
1.155
0.102
0.177
0.969
0.1066
0.022
0.028
0.233
0.674
0.288
0.0665
0.0971
1.1193
5.330
88.132
5.004
0.803
0.167
0.694
Table 3.3: Integration Results of Peaks at 25 Minutes of Steady State Period
%SO2Aso0 2
Mrso2 = PN 2 VN 2 %
2
%N2
rhso2
=
cm 3
0.102
A 2
AN2
64-59
1.9458 3 x 569x 88----2x m
min
88.132 28
4.933 x 10
s
Again, it is necessary to account for the air or else risk grossly underestimating the
sulfur dioxide produced. As expected, the amount of air in the system, which can be
roughly estimated from the absolute area of the oxygen peak, is less than the earlier
run, as the reference gas and the reduction products have continued to fill the collection vessels while the percentage of the air in the collection vessel should generally
decrease over time.
1
Area%0 from SO3
Area%so = 0.051
Area%o2 from air = Area% 0 2 - Area%o2 from SO3 = 5.279
Area% N2 from air = Area%N2
air infrom= 5.279 x 7084 = 19.678
2 afrom
% ir 02 in air
20.9476
55
Area%N2 from reference = Area%N2 - Area%N2 from air = 68.454
Now that the actual relative amount of nitrogen in from the reference gas has been
found, the original calculation for the SO2 mass rate can be recalculated.
rhso2 = 1.9458
M3
Cg
m3
min
x 569-
0.102
x 68.4
68.454
64-59
x-
28
= 6.285x 1
s
Averaging the two steady state results, the steady state production of sulfur dioxide
is 9.208E-5±0.0001.
3.2.2
Gas Chromatograph Analysis as a Function of Temper-
ature
The time dependent analysis used earlier for the mass balances can be performed for
the chromatograph samples also. A sample was taken when the temperature of the
system was 5950 C. Figure 3-10 shows the peak analysis of the sample.
25uV
10000
8000
6000
4000
2000
0
1
2
2
4
6
3
Figure 3-10: Chromatograph Analysis at 595°C
56
7
min
A clear sulfur dioxide peak is present, although it is small due to the low operating temperature. Figure 3-11 shows the presence of the peak at 2.03 seconds, which
has been the characteristic time of the sulfur dioxide peak throughout the experiment.
25UV
200180
16014012010080-
2
60_
401.95
~~~~~~~~~~~~..
. . - - -- - - - - - - .-- - -__
2.00
--
--
2.05
-- -- - . -.
_
- -
. -
. .
.
2.10
. ....
...
..
2.15
2.20
min
Figure 3-11: SO2 Peak at 595°C
The autointegration feature in Chemstation was used to produce the results in Table
3.4. The sulfur dioxide peak is relatively smaller than earlier, which was expected
due to the low operating temperature.
Note the continued decrease of the oxygen
peak from earlier runs. The decrease was earlier predicted as the air in the sample
loop exited the system due to the mass flow of the exit stream and the nitrogen.
Peak
Time
Area
Height
Width
1
2
3
4
5
6
7
0.851
2.032
2.531
3.025
3.724
3.89
6.036
743.7
25.3
9.5
129.3
2721.7
67228.4
2476.1
105.5
4.2
1.3
71.8
601
10109.4
27.1
0.0962
0.1251
0.1175
0.029
0.0708
0.0996
1.0899
Area % Symmetry
1.014
0.035
0.013
0.176
3.711
91.674
3.376
0.181
3.645
1.06
0.348
0.949
0.286
0.656
Table 3.4: Integration Results of Peaks at 595°C
Performing the same exact analysis as earlier, the mass flow rate of the sulfur dioxide
production can be calculated.
57
%s02 ASo0 2
mso2 = PN2VN2 %2
%N2
A
2
AN2
Area% 0 2 from SO3 = 1Area%so 2 = 0.0175
Area%o0 2 from air = Area% 2 - Area%o2 from SOs = 3.694
Area%N2 from air = Area%o2 from air
%N2 in
air = 3.694 x
= 19.678
in ir =0 694 20.9476
Area%N2 from reference= Area%N2 - Area%N 2 from air = 77.906
mso 0 2 =
kg
cm3
0.035
x 575x 77-90
m3
min
77.906
1.9458
6415
= 1.895 x 1028
The production at 5950 C can be plotted on a curve with the production at steady
state, and a production curve can be extrapolated. Figure 3-12 shows the maximum
production possible, which was derived from the difference in mass flow rates found
from the mass balance.
58
x 10 4
18
Unaccounted Mass Rate as a Function of Temperature
I
I
I
I
I
I
16
14
12
Maximum Production, .01159-e 1 16620E-6T
Actual Production, ln(O.99974+4.74956e-7*T)
10
C
0
CU
8
co
6
2
4
2
~~~~~~~~~~~_.---
0
_-
500
I
I
I
I
I
I
550
600
650
700
750
800
Temperature
I
850
900
950
C)
Figure 3-12: SO 2 Production
3.3 Efficiency Calculations
A molecular conversion efficiency resembling a second law efficiency is proposed.
= 7nSO2,actual
(3.2)
nso 2 ,max
Using the results in Figure 3-12, it is possible to plot the efficiency as a function of
temperature. Figure 3-13 shows the relation of efficiency to temperature. Note the 0
efficiency crossing around 550°C. The zero crossing in other studies concerning sulfur
dioxide production is generally at a higher temperature, which can be explained by
the time lag of the temperature in the system relative to the source of the sulfur
dioxide in the collection vessel. The measured sulfur dioxide in the sample vessel is
59
actually a time based average of earlier samples. However, the general trend is clear
and shows the catalytic ability of the 800H + 5%Pt material.
Conversion Efficiency as a Function of Temperature
^
4^
0
C
a)
tl=
550
600
650
700
750
Temperature
800
850
900
950
(C)
Figure 3-13: SO2 Production
Efficiency
It is important to understand the great number of variables in the efficiency calculation. The temperature of the test, catalyst area, and residence time are few of the
factors that affect the efficiency. Even the ability of the system to capture the undesired products in the collection vessel has significant effects on the system efficiency.
Although the system parameters other than temperature have not been factored into
the efficiency calculation, they will be presented here to facilitate future comparative
work.
The sample of 800H + 5%Pt used was 3.325 cm x 20.64 cm x 0.0058 cm. The sample
was cut into 9 separate pieces, which created a total surface area of 138.12 cm2. The
60
sample after use in the experimental loop is show in Figure 3-14.
Figure 3-14: Catalyst Sample After Experiment
3.4
Discussion and Conclusions
The gas chromatograph analysis of the sample streams show significant corrosion.
The most telling data is the molecular conversion efficiency of the system, which
ranged from 0% at approximately 550°C to approximately 10% at 9000 C.
As mentioned earlier, the efficiency presented in this study is merely an approximation, and refinements can be made to the system setup in order to more accurately
find the efficiencies. Improvements can be made by using a printed circuit heat exchanger, which will increase the available surface area and thermal efficiency. Larger
reaction vessels will regulate the pressure of the system such that there are not as
dramatic fluctuations over the course of the run, and also provide greater surface area
and residence time. Efficiency gains can also be achieved by raising the temperature
of the reaction.
In terms of measurement, a large amount of nitrogen was used as a means of carrying
the SO2 into the gas chromatograph. From the chromatograph analysis, it is clear
that the nitrogen used was more than necessary, and scaling back both the flow rate
and the pressure of the nitrogen may actually aid the transport of the SO2.
61
Overall, the research confirmed the catalytic activity of the alloyed material, which
was the desired answer at this stage of the catalytic characterization. The molecular
conversion efficiency of the system reached approximately 10% at 9000 C, which was
an encouraging figure given the loose packing and other possible losses in the system.
This value could be increased in the future as improvements are made both to the
system configuration and measurement techniques.
62
Bibliography
[1] Charles W. Forsberg and Paul S. Pickard. The advanced high-temperature reactor:
Matching nuclear energy systems to thermochemical hydrogen production.
In
American Institute of ChemicalEngineersSpring National Meeting, March 2002.
[2] Lloyd C. Brown, Ryan D. Lentsch, Gottfried E. Besenbruch, Kenneth R. Schultz,
and James E. Funk. Alternative flowsheets for the sulfur-iodine thermochemical
hydrogen cycle. General Atomics Project 30171, 2003.
[3] David Rigual. Characterization of platinum based alloys. Master's project, Massachusetts Institute of Technology, Nuclear Engineering Department, May 2005.
[4] Dr. Robert E. Uhrig. Engineering challenges of the hydrogen economy. The Bent
of Tau Beta Pi, 2005.
[5] Paul M. Mathias and Lloyd C. Brown. Thermodynamics of the sulfur-iodine
cycle for thermochemical hydrogen production. In Annual Meeting of the Society
of ChemicalEngineers,Japan, March 2003.
[6] Ai-Quoc Pham. High efficiency steam electrolyzer. Proceedings of the 2000 DOE
Hydrogen Program Review, 2000.
[7] Daniel M. Ginosar. Catalytic decomposition of sulfuric acid for thermochemical
water splitting processes. Novel Hydrogen Generation Processes (Nuclear and
Solar), January 2005.
63